The aim of the proposed study is to systematically develop and test a dynamic, social network model of utilization and compliance applicable to the problems and the severely mentally ill. Drawing from early insights on the "illness career" approach, dominant model of utilization and compliance, and recent work on the importance of networks and social support for mental health, a theoretical framework is proposed which posits social networks as the mechanism through which individuals recognize and evaluate mental problems, determine when and where to seek professional medical help, and decide whether or not to comply with medically prescribed treatment regimens. Briefly stated, the structure (e.g., the number, strength, type of ties) and function (e.g., advice, material aid, emotional support) interact with cultural content (e.g., attitudes and beliefs toward illness and medical care) to influence critical decisions that individuals make through the onset and course of a mental health problem. Three tasks to be accomplished in the study are: 1) the development of the network- episode model; 2) the modification of traditional survey sample designs and instruments to capture dynamic aspects of utilization on compliance for such problems; and 3) the evaluation and application of four different analytic techniques potentially useful in evaluating the overall explanatory power of this approach and in empirically testing the model's hypotheses. The explanatory study proposed here employs a three-wave prospective panel design to follow first admission patients with major depression and schizophrenia in two mental health agencies (a CMHC and a private outpatient hospital clinic). Semi-structured interviews will be conducted with both clients and the network ties. If the results are promising, this study provides a model and an agenda for mounting larger studies on these or other diagnostic groups. By understanding how the influence of seemingly immutable socio-demographic characteristics posited in traditional utilization and compliance models are in reality proxies for the influence of social networks, this study can inform mental health agencies on strategies for improving out-reach and compliance goals.